R Notebook

library(ggplot2)
library(tidyverse)
library(Seurat)
library(miscTools)
library(viridis)
library(venn)

Read in mutation table

mutation_table <- read.csv(file = paste(dir, "Data_input/geneCellMutationCounts_2.27.19.csv", sep = ""), header = T, row.names = 1)
head(mutation_table)

length(intersect(rownames(mutation_table), rownames(tiss_subset_tumor@meta.data)))

[1] 3620

#drop X.1 and X.2 to clean mutation table
mutation_table$X.1 <- NULL
mutation_table$X.2 <- NULL

Normalize the number of mutations by the raw # of transcripts found

# correct colnames
colnames(mutation_table) <- gsub(pattern = "\\.", replacement = "-", x = colnames(mutation_table))
# Subset table to tumor cells alone and 
cells.to.use.tumor <- tiss_subset_tumor@meta.data$cell_id
mutation_table <- mutation_table[cells.to.use.tumor, ]
# save the names of the mutated genes
genes.to.use.mutations <- colnames(mutation_table)
length(genes.to.use.mutations)

[1] 14397

# set raw counts equal to the genes in the mutation table and subset to only tumor cells
raw_counts <- as.data.frame(as.matrix(tiss_subset_tumor@raw.data))[genes.to.use.mutations, cells.to.use.tumor]
# transpose the raw counts so that colnames are row ids and rownames are cell ids (same as the mutation table)
raw_counts1 <- as.data.frame(t(raw_counts))
dim(raw_counts1)

[1]  3620 14397

# normalize the mutation table by the raw counts
normalized_mutation_table <- mutation_table/raw_counts1
# Change NANs, NAs or Infs to 0
is.nan.data.frame <- function(x)
do.call(cbind, lapply(x, is.nan))
is.infinite.data.frame <- function(x)
do.call(cbind, lapply(x, is.infinite))
normalized_mutation_table[is.nan(normalized_mutation_table)] <- 0
normalized_mutation_table[is.na(normalized_mutation_table)] <- 0
normalized_mutation_table[is.infinite(normalized_mutation_table)] <- 0

Total Mutations Analysis

# Set Table to Report Only the Total Number of Mutations
total_mutations_table <- as.data.frame(rowSums(normalized_mutation_table))
head(total_mutations_table)

total_mutations_table$cell_name <- rownames(total_mutations_table)
colnames(total_mutations_table) <- c('total_mutations_score','cell_id')
head(total_mutations_table)

Add Metadata to normalized mutation table

total_mutations_table$cell_id <- rownames(total_mutations_table)
total_mutations_table_wmetadata <- left_join(total_mutations_table, tiss_subset_tumor@meta.data, by = 'cell_id')
rownames(total_mutations_table_wmetadata) <- total_mutations_table_wmetadata$cell_id

Total Mutations detected by Group

Add total mutation score to metadata

Plot